Sei sulla pagina 1di 6

International Journal of Modern Communication Technologies & Research (IJMCTR)

ISSN: 2321-0850, Volume-2, Issue-2, February 2014

DC Motor Speed Control using Artificial Neural


Network
Yogesh, Swati Gupta, Mahesh Garg
synaptic connections that exist between the neurons. This is
Abstract This paper presents an insight into the speed control true of ANNs as well.
of D.C motor using a Artificial neural network controller to There are a number of articles that use ANNs applications
meet the desired speed. The Neural Network scheme consists of to identify the mathematical DC motor model. Then, this
two parts: one is the neural estimator, which is used to estimate model is applied to control the motor speed. The inverting
the motor speed and the other is the neural controller, which is
forward ANN with two input parameters for adaptive control
used to generate a control signal for a converter. These two
neural networks are trained by feed forward neural network
of DC motor ANNs are applied broadly because all the ANN
algorithm. Standard three layer feed forward neural network signal are transmitted in one direction, the same as in
with sigmoid activation functions in the input and hidden layers automatically control system, the ability of ANNs to learn the
and purelin in the output layer is used. Simulation results are sample,
presented to demonstrate the effectiveness and advantage of From the very beginning, it has been realized by systems
control system of DC motor with ANNs in comparison with the theorists that most real world dynamical systems are
conventional control scheme. For the comparison we used PID nonlinear. However, linearization's of such systems around
control. The purpose of this study is to control the speed of the equilibrium states yield linear models, which are
direct current (DC) motor with Artificial Neural Network
mathematically obedient. In particular, based on the
(ANN) controller using MATLAB application. The Artificial
Neural Network Controller will be design and must be tune, so
superposition principle, the output of the system can be
the comparison between simulation result and experimental computed for any arbitrary input, and alternately, in control
result can be made. The scopes includes the simulation and problems, the input, which optimizes the output in some
modelling of direct current (DC) motor, implementation of sense, can also be determined with relative ease. In most of
Artificial Neural Network Controller into actual DC motor and the adaptive control problems, where the plant parameters are
comparison of MATLAB simulation result with the assumed to be unknown, the fact that the latter occur linearly
experimental result. This research was about introducing the makes the estimation procedure straightforward. The fact that
new ability of in estimating speed and controlling the self excited most nonlinear systems thus far could be approximated
DC motor. In this project, ANN Controller will be used to
satisfactorily by linear models in their normal ranges of
control the speed of DC motor. The ANN Controller will be
programmed to control the speed of DC motor at certain speed
operation has made them attractive in practical contexts as
level. The data from ANN Controller is sent to the DC motor well. It is this combined effect of ease of analysis and
through an interface circuit or a medium called DAQ card. The practical applicability that accounts for the great success of
sensor will be used to detect the speed of motor. Then, the result linear models and has made them the subject of intensive
from sensor is fed back to ANN Controller to find the study for over four decades. In recent years, a rapidly
comparison between the desired output and measured output. advancing technology and a competitive market have required
systems to operate in many cases in regions in the state space
Index Terms DC motor, MATLAB, DAQ card, ANN where linear approximations are no longer satisfactory. To
Controller. cope with such nonlinear problems, research has been
underway on their identification and control using artificial
I. INTRODUCTION neural networks based entirely on measured inputs and
Nowadays, the field of electrical power system control in outputs.
general and motor control in particular has been researching
broadly. The new technologies are applied to these in order to Problem Statement
design the complicated technology system. One of these new When commerce with DC motor, the problem come across
technologies is Artificial Neural Network (ANNs) which with it are efficiency and losses. In order for DC motor to
based on the operating principle of human being nerve neural. function efficiently on a job, it must have some special
It is composed of a large number of highly interconnected controller with it. Thus, the Artificial Neural Network
processing elements (neurons) working in unison to solve Controller will be used. There are too many types of
specific problems. ANNs, like people, learn by example. An controller now a days, but ANN Controller is chosen to
ANN is configured for a specific application, such as pattern interface with the DC motor because in ANN, Non-adaptive
recognition or data classification, through a learning process. control systems have fixed parameters that are used to control
Learning in biological systems involves adjustments to the a system. These types of controllers have proven to be very
successful in controlling linear, or almost linear, systems.

Problems encountered and solutions


Manuscript received Feb. 20, 2014.
Yogesh, M.TECH STUDENT Electrical Engineering Department (RIET
JAIPUR), 9671740921
Problem encountered:-
Swati Gupta, (A.P) Electrical Engineering Department (RIET JAIPUR),
Mahesh Garg, (A.P) Electrical Engineering Department (RIET
i) Control of DC motor speed;
JAIPUR), 9694500083

19 www.erpublication.org
DC Motor Speed Control using Artificial Neural Network

ii)Interface of DC motor with software response has been achieved using the adaptive leaning rate
(MATLAB/SIMULINK); feature in the ANN based controller Also it can be observed
iii) To acquire data from the DC motor that the speed overshooting of the ANN based controller is
significantly lower than the other controllers. This critically
Solutions:- damped speed response has been achieved using the adaptive
leaning rate feature in the ANN based controller.
i) Use of ANN controller to the system;
ii) Implementation of DAQ card to the control board;
iii) Use of encoder from the DC motor to the control board;

Objectives
The objective of the Artificial Neural Network Controller
Design for DC motor using MATLAB an application is it
must control the speed of DC motor with Artificial Neural
Network controller using MATLAB application which the
design of the ANN controller is provided and can be tune.
Each of the experimental result must be compared to the result
of simulation, as a way to attain the closely approximation
value that can be achieved in this system.

II. SIMULATION AND RESULT

Fig. Simulink Block of separately excited DC Motor

Fig-simulink block diagram of Artificial Neural Network


and PID Controller

III. RESULT
SPEED
Fig. - Simulation block of DC Motor With PID Controller

Fig.- Simulation block of DC Motor With ANN Controller


Following diagram is show that the DC Motor simulink Time(s)
diagram with ANN and PID Controller. To varying the
Fig.1- response of speed of DC Motor with PID Controller
different parameter we find the response and compare to
improve settling time with Adjusting gain value with The above figure show the response of speed of DC Motor
respective controller, Also it can be observed that the speed without using Artificial Neural Network .tantalize the better
overshooting of the ANN based controller is significantly result we need to apply a controller as Artificial Neural
lower than the other controllers. This critically damped speed

20 www.erpublication.org
International Journal of Modern Communication Technologies & Research (IJMCTR)
ISSN: 2321-0850, Volume-2, Issue-2, February 2014
Network. to design a controller we used a simple feed
forward neural network with reference signal.

SPEED

Time(s)

Fig4- response of speed of DC Motor using PID with


Tm3=30%Tm

Time(s)
Fig.2- response of speed of DC Motor with PID Controller
with Tm1=75%Tm

The above figure shows the response of speed of DC Motor


without using Artificial Neural Network .with sully voltage to
analyse the better result we need to apply a controller as
Artificial Neural Network. To design a controller we used a
simple feed forward neural network
SPEED SPEED
Time(s)

Fig.- response of speed of DC Motor using ANN


SPEED

Time(s)
Fig3.- response of speed of DC Motor using PID with
Tm2=50%Tm
SPEED

Time(s)

21 www.erpublication.org
DC Motor Speed Control using Artificial Neural Network

Fig5.- response of speed of DC Motor using ANN Controller


with Tm1=75%Tm

Time(s)
Fig8- response of speed of DC Motor using ANN and PID
From the above figure we can say that the response of DC
Motor speed using ANN is better than PID Controller
Time (s)
Response
Fig6- response of speed of DC Motor using ANN with SPEED
Tm2=50%Tm

From the above figure we can say that the response of DC


Motor speed using ANN
SPEED

Fig9

Time(s)
Fig7.- response of speed of DC Motor using ANN with
Tm3=30%Tm

Response with PID


Response with ANN
SPEED

PID Response with ANN

22 www.erpublication.org
International Journal of Modern Communication Technologies & Research (IJMCTR)
ISSN: 2321-0850, Volume-2, Issue-2, February 2014
Fig10.- response of speed of DC Motor using ANN and PID proved that the proposed ANN based controller has a good
with Tm1=75%Tm ability to control the speed of the Separately excited dc
motor, which shows the non-linearity behavior. Experimental
SPEED results verify that this ANN and PID controllers both are
Response with PID controlled of speed of DC Motor with comparatively result.
Response with ANN The tracing error to less that . We can come to a conclusion
Fig.- response of speed of DC Motor using ANN and PID that the proposed artificial neural network based adaptive
with Tm2=50%Tm controller is clearly superior, particularly in the case of
non-linearities, parameter variations and load disturbances.
SPEED The on-line weights and biases updating feature of the ANN
Response with PID can compensate for both parameter changes and disturbances
Response with ANN during operation

FUTURE SCOPE
While the research reported in this thesis shows that an ANN
based adaptive controller performance is superior it still lacks
with some limitations, which provides room for improvement.
Such possible improvements are indicated below, as possible
directions for further work.
In the present work the number of hidden layers and the
number of neurons in the hidden layer are chosen by trial and
error, bearing in mind that the smaller the number, the better it
is in terms of both memory and time taken to implement the
ANN. Further research can be done to find the optimum
number of hidden layers and number of neurons in the hidden
layer. weights and biases updating feature of the ANN can
compensate for both parameter changes and disturbances
during operation. The uses of the adaptive learning rate in the
proposed controller reduce the possibility of overshooting
particularly during the transient conditions. The feedback
provision in the modified ANN motor structure also enhances
the stability of the system.

Fig11.- response of speed of DC Motor using ANN and PID Appendix


with Tm3=30%Tm Calculation
Parameter
IV. CONCLUSION P= 5HP,
The DC motor has been successfully controlled using an V= 240V,
ANN. Two ANNs are trained to emulate functions: estimating Speed=1750 RPM,
the speed of DC motor and controlling the DC motor, Field voltage =150V,
Therefore, and so ANN can replace sensors speed in the J=0.02215 Nm2,
model of the control systems. Using ANN, we dont have to KF=1.976 NmA-1,
calculate the parameters of the motor when designing the B=0.002953Nms,
system control. It is shown an appreciable advantage of Ra=11,2 ,
control system using ANNs, when parameters of the DC La=0.1215 H
motor is variable during the operation of the motors. The 1/B=0.0892
satisfied ability of the system control with ANNs . ANN 1/Ra=0.0108
application can be used in adaptive controlling in the control Tm=J/B
system machine with complicated load. Nowadays, in order to Tm=7.5008
implement the control systems using ANNs for DC motor on Ta=L/Ra
actual hardware, the ANN micro processor is being used. Ta=0.02953
Artificial Neural Network was used as a trainable
non-linear mapping system. The speed of a self exited dc REFERENCES
motor was controlled using the proposed ANN based adaptive [1] Astrom, K. J. and B. Wittenmark, Adaptive Control, Addison-Wesley,
controller. The details of development of the proposed Reading, MA, 1995.
controller were presented, including all analytical derivations. [2] El-khouly, F. M., A. S. Abdel-Gaffer, A. A. Mohammed, and A. M.
Programming and implementation details including hardware Sharaf, Artificial intelligent speed control strategies for permanent
magnet dc motor drives, in Proc. IEEE-IAS Annu. Meeting, 1994, vol.
interfacing were given as well, for both the computer setup 1, pp. 379384. [00345476
and the physical experimentation. [3] Fukuda, T.; Shibata, T., Theory and applications of neural networks for
To controlled speed of DC Motor we used PID Controller for industrial control systems Industrial Electronics, IEEE Transactions
tuning the ANN to improve accuracy of speed. During the on , Volume: 39 Issue: 6 , Dec 1992 pp 472 489 [00170966]
[4] Hoque, M.A., M.R. Zaman, and M.A. Rahman, Artificial neural
experimentation and after observing the results it has been network based controller for permanent magnet dc motor drives, in

23 www.erpublication.org
DC Motor Speed Control using Artificial Neural Network

Proc. IEEE-IAS Annu. Meeting, 1995, vol. 2, pp. 17751780.


[0530521]
[5] Hoque, M.A., M. R. Zaman, and M. A. Rahman, Artificial neural
network based permanent magnet dc motor drives, in Proc. IEEE-IAS
Annu. Meeting, 1995, vol. 1, pp. 98103 [00530289 ]
[6] Levin, A.U.; Narendra, K.S.; Control of nonlinear dynamical systems
using neural networks: controllability and stabilization Neural
Networks, IEEE Transactions on , Volume: 4 Issue: 2 , Mar 1993 pp
192 206 [00207608]
[7] MATLAB Inc. http:///www.mathwork.com
[8] Narendra, K.S.; Parthasarathy, K Identification and control of
dynamical systems using neural networks Neural Networks, IEEE
Transactions on, Volume: 1 Issue: 1, Mar 1990 pp 4 27 [00080202]
[9] Narendra, K.S., Neural networks for control theory and practice
Proceedings of the IEEE , Volume: 84 Issue: 10 , Oct 1996 pp 1385
1406
[10] Narendra, K.S. and S. Mukhopadhyay, Adaptive control using neural
networks and approximate models Neural Networks, IEEE
Transaction on Volume: 8 Issue: 3 , May 1997. [00572089]
[11] Narendra, K.S., Neural networks for real-time control Proceedings
Conference on Decision & Control, San Diego, California
of the 36th
USA, December 1997. [00657581]
[12] Rubaai, A. and R. Kotaru, Online identification and control of a DC
motor using learning adaptation of neural networks Industry
Applications, IEEE Transactions on Volume: 36 Issue: 3, May-June
2000. [00845075]
[13] Murray, R.M., Li,Z., Sastry,S.S., A Mathematical Introduction to
Robotic Manipulation.
[14] Shah,N.N., Kotwal,C.D., The state space modeling of single, two and
three area ALFC
[15] power System using Integral Control and Optimal LQR Control
method, IOSR Journal of Engineering, Mar 2012, Vol 2(3),
pp:501-510.
[16] Mahalanabis, A.K., Kothari, D.P., and Ahson, S.I.: Computer aided
power system analysis and control, Tata McGrawHill, 1988
[17] Haykin, H, Neural Networks: A Comprehensive Foundation.
Piscataway,NJ: IEEE Press, 1994.
[18] De Mel, W.R. and A. N. Poo, Real-Time Control using xPC-Target in
MATLAB International Symposium on Dynamics and Control,
Hanoi, Vietnam, September, 2003. (Submitted)

24 www.erpublication.org

Potrebbero piacerti anche